INDEX

Explanations

instances of dates and time-related phrases

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

WW

-0.16

 WWII

-0.15

-grow

-0.14

Bek

-0.14

ulin

-0.14

.ov

-0.14

piar

-0.14

Û±Û¹Ûµ

-0.14

¬ģ

-0.14

chang

-0.13

POSITIVE LOGITS

0.74

0.65

0.64

Û²Û°Û°

0.43

0.39

Û±Û¹Û¹

0.36

Û²Û°Û±

0.35

 Bush

0.34

0.33

 bush

0.30

Activations Density 0.465%

instances of dates and time-related phrases

No Comments

No Known Activations